Tumor control probability models for nonuniform dose distributions

Tumor control probability models for nonuniform dose distributions are reviewed, including theoretical predictions of partial volume effects and implications for future treatment strategies. In vitro measurements support the assumption that the most resistant cells in a tumor typically have SF2(surviving fraction at 2 Gy) values in the range 0.4-0.6. This supports the following hypothesis: a boost to a partial volume of a tumor of 10-15 Gy would make negligible the likelihood of a recurrence in the boost region only. Diminishing returns are predicted for boosts larger than about 10-15 Gy more than the planning target volume (PTV) minimum dose (unless the volume associated with the minimum dose is very small). The expected benefit of such boosts increases with the fraction of the PTV volume which is boosted. The importance of this fact for conformal radiotherapy is that, for some sites with nearby dose-limiting normal tissues (such as prostate), intensity modulated, or segmental field, techniques could be used to maximize the volume of such partial PTV boosts. Tumor control probability benefit from such boosts is predicted across a wide range of possible radiobiological parameters. For a patient whose most resistant cells have an SF2 equal to 0.5 Gy, a partial volume boost of greater than or similar to 10-15 Gy is predicted to have the same effect as a uniform increase in dose of Delta D similar or equal to 2 Gy \log(2)(1 - v(BOOST))\, where v(BOOST) is the relative boost volume. Partial volume boosts are a theoretically attractive way of increasing TCP when the entire gross target volume cannot be raised to the highest allowed dose.